This proposal concerns the search for optimal scintillator crystals for positron emission tomography (PET). At present, the availability and performance of PET systems is limited chiefly by the unavailability of scintillators with sufficiently short attenuation lengths, high luminosities, short decay times, and low cost. We propose to undertake a systematic and exhaustive search to find optimal PET scintillator materials, involving a combination of computational and experimental screening of the entire NIST Crystal Data database. First, the software tools, developed during the course of a previous SBIR project, calculate the attenuation lengths, photoelectric fractions, densities, and costs for all known materials and prioritize them according to moieties of a set of PET figures of merit developed by experts in the field. Next, the experimental tools, developed at Crystal Research, LBNL, and ALEM Associates, measure the luminosities, decay times, emission wavelengths, and ease of crystal growth of the most promising candidates, using small powder samples. Finally, the software calculates the set of complete PET figures of merit for all the compounds studied. The result is expected to be a short list of compounds having the highest likelihood of optimal PET performance and therefore suitable for single crystal studies during Phase II. PROPOSED COMMERCIAL APPLICATION: NOT AVAILABLE